Use the Number of Passengers on and off Backstepping the OD Matrix Methods Research

In order to reduce the waste of human and material resources in the actual OD survey and to make the investigation more convenient, the following investigation was conducted on behavioral characteristics of passengers who traveled by bus using analysis from the basic distribution of the distance of bus passengers traveled. Based on the probability of bus passengers would get off, the number of passengers on and off as the conditions already know, proposed a single bus line passenger OD matrix inverse method in accordance with the nature of land use on the site together, and then extended to the public transportation network. This achieved the association of OD matrix of line to region. Finally, determination of the traffic zone OD matrix by division of the traffic zone status through case study, verified the feasibility and effectiveness. OD backstepping technique originated in the motor vehicle OD matrix inverse, inverse OD is based on observations of traffic flow from the status of road network inverse the status of traffic partition OD matrix. A Commonly used models of estimate for the OD matrix are the maximum entropy model (Chen 1993), generalized minimum least square model, information minimal model, maximum likelihood model(Wang 1993), and so on. Structural models can also be applied to calculate the transit passenger OD (Yin 2000). The inquiry process mainly through the analysis of measured data or results, is used to calibrate the distribution model, while using the minimum deviation of the calculated and measured value of the upper and lower number of passengers as the objective function, then obtained the OD traffic matrix by using the distribution model (Ying 2002). This study draws upon previous research used in structural models in order to improve usability of the target and inverse bus trip OD matrix. Progressive approach